Coombs, Clyde H. (Clyde Hamilton), 1912-1988

Founder of the mathematical psychology program at the University of Michigan.

From the description of Clyde Hamilton Coombs papers, 1932-1988, (bulk 1964-1987). (University of Michigan). WorldCat record id: 34420097

Clyde H. Coombs, the founder of the Mathematical Psychology program at the University of Michigan, was born on July 22, 1912. He began his undergraduate work in engineering and mathematics at Santa Barbara State College, but changed his major to psychology and transferred to the University of California-Berkeley. He received his A.B. (1935) and M.S. (1937) from Berkeley, studying with Robert Tryon, Werner Brown, and Nathan Shock. Coombs then studied with and served as a research assistant to L. L. Thurstone at the University of Chicago. He received his Ph.D. from Chicago in 1940. During World War II, Coombs joined the U.S. War Department, serving in the Adjutant General's Office (AGO) in Washington, D.C. His work on the diagnosis and counseling of demobilized soldiers won him the Legion of Merit in 1945.

After the war, Coombs continued to work with the AGO, rising to the rank of major. Preferring a research life, Coombs returned to an academic setting in 1947 at the fast-growing Psychology Department at the University of Michigan. This was the beginning of a professional appointment which would continue beyond his nominal "retirement" in 1983. Coombs' superb abilities as a teacher, enormous productivity as a researcher and scholar, and his administrative capabilities served him well as he advanced to the rank of associate professor in 1949, full professor in 1953, and professor emeritus of psychology upon his retirement. Throughout his long career at the University of Michigan, Coombs travelled widely, lectured, and was a visiting scholar at several schools.

It was during his first such academic leave at Harvard University in 1948-1949 that Coombs began to bring together his interests in mathematics and psychology in the measurement of subjective phenomena. He concluded that psychological measurements could be thought of in terms of two spatial relationships, dominance and proximity, and set theory. He also noted that psychologists chose their model of measurement based on their subjective theoretical ideas rather than holding that measurements were objective data. These ideas on the dynamics of data, measurement, and subjectivity dominated Coombs'research and teaching during the 1950s as he produced several articles and designed seminars to further the research. In this he epitomized the researcher as teacher. Coombs so quickly incorporated the results of his experiments into his teaching that the texts for the courses consisted of mimeographs of his latest work and thought. Eventually Coombs pulled these ideas together in his seminal work, A Theory of Data (1964), completed after a year's fellowship at the Center for Advanced Study in the Behavioral Sciences in Palo Alto.

Coombs' work with mathematics, measurement, and psychology was rewarded with the founding of the Michigan Mathematical Psychology Program in the early 1960s. This institution provided a framework in which Coombs could do his research and continue to attract graduate students interested in understanding human behavior quantitatively. During this time Coombs developed two concepts on preference and decision making, "unfolding theory" and "portfolio theory" which were closely linked to his earlier idea that measurement could be reduced to orders or proximities. In addition to his usual load of teaching, research and writing, and administration, Coombs collaborated with former students Amos Tversky and Robyn Dawes on Mathematical Psychology (1970). This work was enthusiastically reviewed and soon came to be the standard text in the field.

Coombs remained very productive throughout his career, publishing over seventy-five articles and four books. His last book, The Structure of Conflict (1988), was published shortly after Coombs' death in February 1988. This book was also a collaborative effort with a former student, George Avrunin. They examined the preference decisions made by two people seeking dissimilar outcomes and noted these decisions mirror behavior of a single person faced with one multivariate alternative. Throughout his career Coombs remained open to new ideas and was quick to lend his experience and expertise to his many students in collaborative efforts to get their work published. These students were appreciative of Coombs' efforts and formed a world-wide network of loyal friends.

Beyond his teaching and research, Coombs served his science and profession tirelessly. His offices and roles included: president of the Society for Mathematical Psychology (1977), president of the Psychometric Society (1955-1956), president of the Division of Measurement and Evaluation of the American Psychological Association (1958-1959), membership on a wide variety of editorial boards, and on many research review committees for the National Science Foundation. His professional honors included being: an honorary fellow of the American Statistical Society (1959), a member of the American Academy of Arts and Sciences (1977), elected to the National Academy of Sciences (1982), a Fulbright (1955-1956) and a Guggenheim fellow (1970), and awarded the Distinguished Scientific Contribution Award by the APA in 1985.

Coombs met his future wife, Lolagene Convis, while both were graduate students at the University of Chicago. They were married and had two sons, Steve and Doug. They shared an interest in travel and co-authored articles on family composition preferences in the 1970s.

From the guide to the Clyde H. Coombs Papers, 1932-1988, 1964-1987, (Bentley Historical Library University of Michigan)